A quarterly marketing review is a 90-minute check of what your clinic's marketing did last quarter, what changed underneath it, and what is no longer compliant. Run it every 90 days. The reason is arithmetic, not habit. Between 6 January and 31 August 2026, Google's own documentation feed logged 35 changes to how Search works, a median of seven days apart, and Google started six named ranking updates in the same window. A clinic website does not usually drift out of compliance because someone edited it. It drifts because the ground moved and nobody looked. The template below is the whole thing: seven sections, 90 minutes, and one decision at the end.

What is a quarterly marketing review, and what is it not?

It is a fixed, repeatable hour and a half, every 90 days, in which one person looks at four things: the numbers, the changes you made, the changes the platforms and regulators made, and the compliance surface. Then they write down one decision.

It is not a report for a board. It is not a strategy day. It is not the same as the 30-minute website audit, which is a one-off sweep of a single site to find what is already broken. The audit answers "is this page wrong today". The quarterly review answers "what has moved since the last time I looked, and does that make something wrong now".

The difference matters because most compliance failures on Australian clinic websites are not authoring errors. They are ageing. A page that was correct in March can be a problem in September without a single character changing.

The one-line version. Your website is a claim you made on a particular day, about rules that were true on that day, on a search engine that behaved a particular way on that day. All three of those things have a shelf life. The review is how you find out which one expired.

Why 90 days, and not once a year?

Because of how often the ground actually moves. We counted it rather than guessed.

Google publishes an RSS feed of changes to its own Search Central documentation, described by Google as "the latest major updates made to the Google Search Central documentation". We pulled that feed on 1 September 2026 and counted every item in it. There were 35 documented changes, running from 6 January 2026 to 31 August 2026. That is 237 days. The median gap between two change dates was seven days. The longest quiet stretch in the whole eight months was 25 days.

Separately, Google's Search Status Dashboard lists named ranking updates. Six of them started inside that same window: the February 2026 Discover update on 5 February, the March 2026 spam update on 24 March, the March 2026 core update on 27 March, the May 2026 core update on 21 May, the June 2026 spam update on 24 June, and the August 2026 spam update on 18 August. Across the full twelve months to 1 September 2026 there were seven, and the longest gap between any two of them was 56 days.

Now turn that into review intervals. The point is not vigilance. It is attribution.

What a single review has to explain, by how often you run one. Rates are the 2026 figures above, scaled to the interval.
Review intervalDocumented Google changes since your last lookNamed ranking updatesCan you tell what moved you?
Monthly.About 4.About 1, often none.Yes, but you are reviewing noise most months.
Quarterly (90 days).About 13.About 2.Usually. Two candidate causes is a question you can answer.
Half-yearly.About 27.About 5.Rarely. Too many candidates, and the traffic chart has healed or hasn't.
Annually.About 54.About 9.No. You will name whichever update you happen to have read about.

That is the whole argument for 90 days. At a quarter you usually have one or two plausible causes for a traffic change and you can go and check them. At a year you have nine ranking updates, fifty-odd documented changes, whatever your competitors did, and whatever your own staff added to the site. Nobody can reason about that. So nobody tries, and the review becomes a slide that says "SEO is down".

What actually changed underneath your site this year?

Here are the 2026 items from that feed that a healthcare practice should care about, with the dates Google published them. This is what section three of the template is looking for.

Selected 2026 entries from Google's Search Central documentation feed, and why a clinic should care. Titles are Google's own wording.
DateGoogle's entryWhy it matters to a practice
13 April 2026. Introducing a new spam policy for "back button hijacking". A new spam policy is a new way to be demoted. Worth knowing your site does not do it.
8 May 2026. "Deprecating the FAQ rich result feature". Google's note says the feature "will no longer appear in Google Search starting May 7, 2026". Thousands of clinic pages still carry FAQ schema built to win a result that no longer exists.
15 May 2026. "Clarifying that spam policies apply to generative AI responses in Google Search". The rules you can be punished under now cover the AI answer box, not just the blue links.
27 May 2026. "Preferred sources is available in AI Mode and AI Overviews". Another surface where a patient can meet your clinic without visiting your site.
15 June 2026. "Removing documentation for the FAQ rich result feature". The documentation is gone, so a staff member checking "how do I do FAQ schema" in September finds nothing.
24 July 2026. "Added a new review snippet guideline". Review rules are the highest-risk surface a clinic has. See section five.
28 August 2026. "Updated the site reputation policy". Relevant to any practice running content on a third-party domain or a partner site.

The FAQ line is the clearest example of why the review exists. FAQ rich results stopped appearing in Google Search on 7 May 2026. Keeping the Q&A content on the page is still worth doing, because AI answer engines read it. Keeping the schema is harmless. But if your agency is still billing you for "FAQ schema to win rich results", that line item bought you nothing after 7 May, and a quarterly review is where you would have caught it in June rather than next year.

On the Australian side, two dates belong in the same list. The AHPRA and National Boards guidelines for non-surgical cosmetic procedures took effect on 2 September 2025, which we covered for nurses and midwives and for cosmetic clinics. And the Therapeutic Goods Administration currently has a consultation open on changes to the Therapeutic Goods Advertising Code. The law firm King & Wood Mallesons, writing on 14 July 2026, states that "Responses may be submitted on the consultation page until 3 September 2026". Its proposals include advertising rules for software-based and AI-based medical devices, and whether paid influencer testimonials should be allowed for therapeutic sunscreens.

One honesty note on that TGA date. The TGA's own consultation page did not respond to repeated requests from us on 1 September 2026, so the closing date above is quoted from King & Wood Mallesons, not read off the TGA site. Check tga.gov.au before you rely on it. That is exactly the kind of check a review is for.

The template: what to check, in what order

Copy this. Seven sections, roughly 90 minutes, run on the first working day of each quarter. The order matters: numbers first so you are not reasoning from memory, causes second, compliance third, decision last.

The quarterly marketing review template for an Australian healthcare practice.
#SectionTimeWhat you actually do
1 The numbers. 10 min Write down four figures for the quarter and the same quarter last year: new patient enquiries, phone calls, Search Console clicks, and Search Console average position for your three money pages. Four numbers, not a dashboard.
2 What you changed. 10 min List every page added or edited, every new practitioner bio, every new service, every widget someone installed. Everything on that list is copy nobody has compliance-checked.
3 What changed underneath you. 15 min Open Google's Search Status Dashboard and its documentation feed. Note any ranking update that started in the quarter and mark it on your traffic chart. Then check the AHPRA news page and the TGA consultations page for anything with a commencement date.
4 The compliance sweep. 25 min Work section 133 of the National Law in order: false or misleading claims, inducements without terms, testimonials, unreasonable expectations of treatment, encouraging unnecessary use. Then titles, registration numbers, before-and-after imagery, and any prescription-only substance named on the site.
5 The register check. 10 min Take every practitioner named anywhere on the site or your socials and look each one up on the public register. Confirm registration is current and the title on the page matches the register. Remove anyone who has left.
6 The reviews and socials pass. 10 min Read the last quarter of Google reviews and every social caption. Look for patient testimonials about clinical care that you have reposted, quoted or replied to in a way that adopts them.
7 One decision. 10 min Write one sentence: the single thing you will change before the next review, and who owns it. Not a list of twelve. One.

Section seven is the part people delete, and it is the part that makes the other six worth doing. A review that ends in a list of twelve actions produces zero. A review that ends in one named owner and one sentence produces one, and one a quarter is four a year.

Which compliance items go stale fastest?

In our experience of auditing Australian healthcare sites, four things age faster than the rest, and all four are invisible from the page you wrote.

Practitioner titles and registration status. People leave. Registration lapses. A page that correctly called someone a registered nurse in March is a title problem in September if they are no longer on the register. Nothing on your site changed. The register did. This is why section five of the template is a register check and not a proofread.

Reviews you did not write. Testimonials about clinical care are prohibited under section 133 of the National Law, and AHPRA's position is that this covers material you control. Every quarter your Google profile fills with new reviews. Some of them will be detailed clinical praise. You did not solicit them, but what you do next matters, and replying "thank you, we are so glad the treatment worked" is you adopting the claim. Google added a new review snippet guideline on 24 July 2026 covering fake and undisclosed incentivised reviews, so a clinic that offers anything in exchange for a review now has a platform problem stacked on top of the AHPRA one.

Anything naming a prescription-only substance. The TGA rules on advertising prescription medicines to the public are unforgiving, and a page written before a product's status or indication changed can go from compliant to not without an edit. We covered the mechanics in advertising prescription medicines.

Offers, prices and inducements. A "$99 new patient special" that ran last spring and is still on a landing page is an inducement whose terms have expired. It is also the single easiest thing for a complainant to screenshot.

If you would rather not run the compliance sweep by hand, that is what the GhostRank audit does, and the AHPRA advertising guidelines hub has the section 133 grounds written out in full.

Who runs it, and how long does it really take?

One person. Ideally the practice manager, not the owner, and not the agency.

Not the agency, because sections two and three ask what the agency did and what it should have told you. That is not a question to outsource to the answer. The agency should be sent the finished review, not asked to write it.

Not the owner, usually, because the review needs to actually happen on a date and clinical owners are the least protected diary in the building. A practice manager with a recurring 90-minute block will run it. A principal dentist will move it three times and then it is February.

On timing: the first one takes longer than 90 minutes, probably closer to three hours, because you are building the baseline. Sections one, two and five are the ones that stretch. From the second quarter onward, 90 minutes is realistic, because you are only looking at the delta.

If you have an agency, add one line to the brief: they send you their own list for section three, in writing, before the review. Any agency worth keeping already tracks Google's changes and will have it ready. An agency that cannot produce it is telling you something useful, and that is roughly the test we described in whether your agency understands AHPRA.

What this review cannot do for you

Four limits, stated plainly.

Ninety days is a judgement, not a finding. The counts above tell you how often things change. They do not prove 90 days is the right interval for your practice. A cosmetic clinic in a contested market and a rural GP practice with two competitors are not running the same risk, and the second one could reasonably review twice a year.

Our Google count is a floor, not a census. Google describes that feed as "major updates" to its documentation. Changes it does not consider major, and changes to ranking that it never documents at all, are not in the 35. The real number is higher and unknowable.

The window is eight months, not a year. The feed we counted ran from 6 January to 31 August 2026. Everything we say about annual rates is that eight-month rate scaled up, and we have said so in the table caption rather than dressing it up as a measured year.

This is general information, not legal advice. A review template is not a compliance opinion, and nothing here is a substitute for advice on your own advertising from someone who has read it. If a matter is already live with AHPRA, stop reading marketing blogs and call a lawyer.

One last thing. The most common outcome of a first quarterly review is not a compliance scare. It is finding three pages nobody remembers publishing, one practitioner who left in April, and an offer that expired in 2025. That is a good result. It means the review is doing the job the calendar was supposed to do.